Welcome, reader, and thanks for being here.
I want to talk to you about No Confidence today, but I also want to talk to you about distributed systems architecture, a profession obscure enough that most people have never even heard of it. (It is relevant, I promise.) I certainly didn’t understand what it was the first, second, third, or even fifteenth time I asked my father what he did for a living.
Obviously I had to go into the profession for myself. How could I not, when even learning what it meant was such a delightful challenge?
It’s probably easier to comprehend today than it was in the 1980s, of course, since now we’re all plugged into the Internet, the biggest distributed system humanity has ever created save capitalism itself. So there’s your clue for the first part of the name: a distributed system is one that is spread out over a wide area, made up of countless individual components all working in concert.
In the industry, “distributed” means something else, as well. It means that there is no single point of failure, no central authority that everything relies on for its operation. Imagine going in to renew your driver’s license, but instead of taking a number and waiting for it to be called, you go to the first window and tell them why you’re here, and then go to a second window and give them your documentation, and so on and so forth like a bureaucratic fast-food drive-thru.
It might seem like it works more smoothly at first, right up until the first time a person can’t explain exactly what they need to the receptionist. Then everyone has to wait on that interaction to finish before anyone else can get seen.
Reader, if you just had the thought, “well, clearly they should have a few receptionists at that desk, because everyone has to go there first anyway,” then you begin to understand the architecture part of the profession, as well. A distributed systems architect must not only know how the system itself works, they must know how people interact with it, too.
The systems part is the one I find both easiest and hardest to comprehend and define, largely because just about anything can be a system, as long as it’s made up of more than one part. The agonizingly long process to get a driver’s license is a system of people and rules; a wall clock is a system of gears and axles; an electric power grid is a system of generators, transformers, and transmission lines. The only particularly defining feature of a system, as opposed to just a collection of individual things, is that changing one part of a system affects the rest.
When a system is healthy, everything works, well, like clockwork. A new number gets called every few minutes, so you can see yours getting closer and you don’t get too impatient; the hour hand hits a new number exactly as the minute hand hits twelve; another generator spins up and starts producing electricity when the demand goes up in the heat of summer.
The more separate pieces a system has, the more impressive it is when they all work together as designed; who among us hasn’t marveled at a perfectly-synchronized marching band or laughed at a ridiculous Rube Goldberg machine?
Of course, it’s that very ridiculousness that hints at the potential trap of systems. Rube Goldberg’s drawings were so funny because those machines were so much more likely to fail than succeed. A single domino out of place, or a mouse jumping the wrong way, and the machine would not just fail to work, it would become utterly useless.
In systems architecture, we call this a “failure state”: the system can no longer perform its designed function just by relying on the nature of the system itself. If each clerk at the motor vehicle office is working on a particularly long and cumbersome transaction, then the people with quick renewals stop cycling through, numbers stop getting called, and everyone in the building gets more impatient. A misaligned hour hand might hit a new number when the minute hand reaches seven, instead of twelve.
Some failure states aren’t so benign. In February 2021, the electric power grid in Texas suffered catastrophic failures that left millions of homes without power for days, resulting in over two hundred deaths, despite everything the state could do to restore power more quickly. Even that, though, was a fraction of how bad it could have been; if power plant operators hadn’t acted quickly enough to save as much of the grid as they did, the entire state could have lost power, potentially for months.
This is not a hypothetical scenario; this was minutes away from happening. Power plant generators are some of the most phenomenally powerful machines that humanity has produced, and like all of our most powerful inventions, they need to be part of a carefully tended and maintained system or all that power will tear it apart from within. The surest (non-violent) way to destroy a power plant generator is to increase its load past what the system can bear; because power plants represent such a sizable financial investment, all power plants are designed to shut themselves off rather than vibrate themselves to pieces, if the load gets too high.
This overload was state-wide, though, which meant that not just one or two of them but every generator would have shut down. The state would have had to perform a “black start” to restore power to the grid, a process rife with points for potential failure in part because it has never been done before.
Some systems get so complex that we’re not even sure if we can get them running again, if they stop. “Too big to fail” does not mean that they can’t fail; it just means that we’re really scared of what will happen if they do.
This is why the discipline of distributed systems architecture exists. We have learned so much as a species, in between the time when electric power grids were first spreading across North America and now. Dependable systems reliability has become essential for so much of what we do now that the majority of our world runs on the Internet, and the intuitions of yestercentury have become the well-established best practices of today.
One thing we’ve learned the hard way (and proved mathematically, for good measure) is that no system lasts forever. Anything with enough pieces can get into a failure state, and diagnosing and fixing failure states is difficult, costly, and sometimes impossible. Every complex system can get into a failure state it can’t get out of on its own, which means that if you leave it running long enough, then eventually, every complex system will.
This, of course, is where the ubiquitous advice of “Have you tried turning it off and back on again?” comes from. Any electronic device is a system; any system can get into a failure state. If the system is simple enough that you can just start it up again from scratch, then that will always be the easiest way to get it working again.
Some systems, of course, aren’t that simple. So let’s talk about No Confidence.
First off: why the name? It comes from the “motion of no confidence,” a feature of parliamentary governments. In these systems, where the parliament elects the government, the parliament can likewise unelect that government once it no longer has their support. In America, it is we the citizens who elect our government, and thus it must be we, the American people, who raise our voices to tell our government that it has lost our confidence. We no longer have faith that our government will actually do right by us, and that lost faith has real repercussions, and it also has a name, recognized and feared in halls of government the world over: No Confidence.
The No Confidence Amendment doesn’t attempt to solve any of the many problems with American politics directly. Movements have been trying to come up with the best solutions to the various problems that plague our society for longer than I’ve been alive, and one or two have been at least mildly successful, but on the whole American politics resists any efforts to nudge it away from the status quo.
No Confidence makes no attempt to solve those problems, because those are not problems that can be solved by a single person. Instead, it simply addresses the failure state itself. Our government is a complex system that has been operating continuously for two hundred fifty years, and the failure state it has gotten into is not one that our existing electoral system is capable of fixing. Because of how rigid American politics has become, it’s only a very small part of our government, the so-called “swing” or “battleground” states and seats, that actually have a chance of changing in each election. Every part of the government that does not change persists the failure state into the new term.
No Confidence allows the American people to demand a true break from the existing system. It creates a single point in time when we change out every political role in our government—that is, the people who make decisions, not the millions of hardworking government employees who actually keep our country running day-to-day, regardless of which party is holding the reins.
When we ratify No Confidence, we strip every single one of those decision-makers of their authority, along with anyone they placed in a persisting role. We hit “undo” or “pause” on every decision that has been made since the current administration took office, until we have someone we can trust to look over them and decide which were good for the country, and which were not. We guarantee every government employee (both those who have been faithfully executing their roles for years and those whose employment was wrongfully terminated by the current administration) their job and their paycheck, for the interim. We trust these selfless functionaries to keep our country running, just like we trust our brave soldiers to keep our country safe, while we the citizens of the United States do the really hard part.
Only the first of the eleven sections of the No Confidence text describe the above: what happens the moment the amendment becomes part of our Constitution (or when we need it again in another two centuries or so). The other ten describe what we do after we clean house: how we elect an Interim Congress to oversee the cleanup efforts, their role in the reconstruction of America, and the process for returning to a more normal (and, hopefully, more healthy) political system once they’re done.
I’ll go into more detail next time, but again, No Confidence isn’t designed to fix any of our problems itself. When we ratify it, all it will do is leave us with a trustworthy interim government that actually wants to make lasting progress in our country—but even if they don’t, we will have given our existing system enough of a break that the next Congress will have much more political freedom than the one we have now. No Confidence is a way to pick ourselves back up out of the dirt, dust ourselves off, and stagger forward to the starting line.
I wrote No Confidence and designed its ratification strategy to be something we could use in the last of the last resorts. This is not an amendment that you would ever write into your Constitution with a healthy government; it’s only politically viable at a point in history when the vast, typically-apolitical majority of Americans are unable to avoid seeing just how bad things have gotten.
That’s on purpose. No Confidence is an extremely powerful disinfectant; if we allow it to be used just when one group of people doesn’t like what the government is doing, then it will itself become a corruptible part of the system, and it will lose its power. This way, we will only ever—we can only ever—reach for it when the situation is, as it is now, so wrong that everyone can see it.
It also means that if we do attempt to change things slowly and piecemeal by means of midterm elections and legislation, or if someone deposes this administration by means of violence, we will be giving up the option for No Confidence for good, unless and until things get this bad again.
There might still be paths forward for America and the world in that case, but I can’t see them, reader, and I gotta tell you, I’m tired of waiting and hoping that maybe someone comes up with some solution someday in the future. This solution exists now, and it doesn’t require waiting, it requires doing. Now, tomorrow, the next day, maybe skip the one after that if you need a break, but all that No Confidence demands is that you decide, right now, that you want the America of tomorrow to be strong, proud, great, and free.
Until next time, stay safe.
Thank you for reading Unfucking The World. We have an enormous task ahead of us, and I’m so glad that I get to start introducing you to the tool I’ve been crafting, refining, and honing to address it.
There will be more to do for No Confidence in the future, but the most important thing is to grow our voice as fast as we can. Share the news on social media with the hashtag #UTW and the URL UTW.vote, and make sure you and everyone you know has subscribed.
Questions, challenges, additions, and thoughtful disagreements are encouraged in the comments; bring your good faith, and others will do the same. Today’s topic: a change on the scale of No Confidence is inherently scary. What worries you the most about it?
Share this with someone you know who wants to end this nightmare, and stay safe.
–Danielle
Este artículo también está disponible en español.
